Don’t know yet. My tinnitus is better controlled.
1 Like
This is my favorite post so far here! You actually bought a REM machine!!! Bravo 
2 weeks follow up.
After playing around I realized some things.
The main conclusion is that, if you use in situ audiogram for your hearing aid, you most likely will be fine without real ear confirmation. It only works with IN SITU AUDIOGRAM for first fit and your manufacturer fitting, not NAL-2. You cannot use your audiologist audiogram for first fit.
Read on for details.
Most Importantly and #1. I used my audiologist audiogram for the real ear measurements. When I used it for first fit, there was often >10 db difference between frequencies over 4 K with all hearing aids. When I used in situ audiograms with each individual hearing aid first fit, only one differed from my audiologist audiogram real ear measurement by more than 3 db. 2 were perfect fits between 4-8K. The only one that still failed >10 db short in the high frequencies was the Signia Silk. I had to go through and use the frequency shaping to match the higher frequencies.
Second. Don’t play around with compression much. I used compression to match perfectly the 50,65, and 80 db curves and everything sounded compressed and vomited out. It was perfectly understandable but unpleasant. Using the gain only to match each level cannot achieve 3 perfect curves, but if you match the 65 db perfectly everything else sounds great. If you have more of a hard time with soft sounds you could base it on the 50 db curve, or compromise, but over compressing sounds absolutely terrible.
In summary, you probably can use your in situ audiogram without real ear measurements with Signia active pro, Signia pure charge and go, Phonak Audeo paradise, and I assume most other aids. If you have Signia Silks, your first fit is probably 10 db low, at least, above 4 k.
Edited: I also use speech mapping and the speech sounds for REM as it’s supposed to be more accurate than just using the default noise settings. However it wasn’t a huge difference.
5 Likes
Are you saying that one either uses the audiogram with REM or starts fresh and does in situ? Then from either of those settings one can tweak things?
Programming the aids with the in situ audiogram gave me closer real ear results measured to my Audiologist audiogram. When using REM, the audiologist audiogram should be the end point because it is more accurate and the end result is absolute. However, if I use my audiologist audiogram to program the hearing aids, they fall way short of that same audiogram on REM. This likely has to do with the in situ audiogram better reflecting the actual hearing aid instrument to ear dynamics. It’s counterintuitive but reproducible on the various models I’ve tried.
I imagine audiologist program their own audiogram into the initial fittings and get terrible results which is why 2/3 of their initial fittings don’t align near REM. The industry first fit process is flawed if they aren’t using the in situ audiogram a for first fit. Companies need to study and emphasize this to hearing instrument specialists and audiologists.
Regarding your question about tweaking. You of course can tweak regardless of initial first fit programming, but will not have to tweak as much if you use in situ audiogram for first fit.
Given the manufacturers are building in situ into the software, they likely have similar data to what you found. This seems to suggests that an audiogram isn’t directly useful for programming aids which is consistent with the need for REM. If in situ can approximate REM, or even surpass it, then REM and possibly audiologists become less necessary and both DIY and OTC could expand bringing costs down. It sounds like the primary challenge with in situ is the subjectivity of the person being fitted.
Were there any clear points of difference between REM and in situ or was it all within the margins of error?
1 Like
The first fit in situ settings were within 5 db of the real ear NAL-2 settings in most frequencies using the manufacturer first fit-algorithm (not nal-2 first fit) for Signia pure charge and go, Signia Active Pro, Phonak Audeo paradise. The high frequencies fell way short on the signia silk (I suspect the manufacurer purposely did this to avoid high frequency feedback in this model.)
You need to be sure to select experienced or power user, and 100% with acclimatization disabled or it will fall short of REM.
REM is still the gold standard.
I have to say that my recent setup was done in with by myself and set with the audiogram+ (resounds settings which is very close to nal2) and it has been terrific. The quality as compared to the settings by my Audi are really much better. I was quite surprised.
I also turn off noise reduction and similar adjusting programs because it bugs me but that’s a personal preference 
Maybe I have a stupid idea, but what about or why modern hearing aids, doesn’t have microphones pointing your ear to measure what your ear is getting, so with that every hearing aid could do REM or at least a very close REM.
I know there’s also the speaker from the computer part, but that part would be way cheaper than current REM systems by far.
Iirc, this topic has come up before: Starkey did just that in the past, but nobody was interested in REM at that time. So it tanked. Search the forum for more info.
Oticon also used to provided audicians with a USB REM device REMsp from MedRX, whenever they sold 4 aids, IIRC. These were cheap. However, I haven’t been able to find or buy one myself.
2 Likes
Thanks for updating me with those, I’m new here so getting that old information back was great.
Hopefully cheap REM could be around the corner.
Most modern high-end HAs provide an onboard hearing ‘test’ that generates tones at various frequencies and allows the audi. to establish an audiogram based on what the HAs are delivering to the eardrum.
I’ve had REM based fittings and I’ve had the average quick fitting. For a number of reasons, I lost faith in the audi. and went down the DIY route using Phonak Target.
Of the three fitting methods, I find that using the HA onboard audiogram within the Phonak Target software gives me a way better fitting than anything I was able to achieve in the audi’s office. It provides many of the benefits of REM as it is based on the sounds actually heard by your own ears by your own HAs.
IMO, for most people, this is the best way for HAs to be managed on an ongoing basis. Having to wait weeks for an audi appointment and being rushed in and out of the office is horrible. This approach does need a DIY element, but if manufacturers were serious about helping consumers rather than audis, they would build the capability into their consumer level app.
OTC devices are allowing consumers to do this, and it will become the norm (hopefully) from the major suppliers also.
2 Likes
What you’re referring to here is in-situ audiometry. It’s like generating the threshold level pure tone test across the frequencies, pretty much the same way your HCP does when they put you inside a sound booth to test and get your audiogram.
Of course REM equipment is too expensive for DIY folks, so in-situ audiometry is the next best thing they can use to verify their HAs’ performance. However, in-situ is not superior to REM for several reasons which I’ll list out below Otherwise, HCPs would have done in-situ audiometry instead of REM for their patients already.
-
In-situ is only a pure tone test in complete silence
-
it only measures the threshold level detection by your brain, nothing else
-
Also, in-situ is a tone generated internally by the hearing aids on command from Genie 2 (in turn on command from you). This results in a bias here if you administer it for yourself, because even if you didn’t hear the threshold sound, YOU KNOW that you pushed that button to activate the pure tone sound. At least in the booth test, you don’t know when the HCP activates the pure tone sound.
-
In-situ does not cover how well the microphones on your hearing aids work to pick up the real sounds because the tone is generated internally inside the HAs.
-
In-situ does not cover real life sound scenarios, albeit it’s still a simulation as the sounds are generated by the speakers in the HCP office.
-
In-situ cannot verify how well your HAs (and the whole setup) amplify against a target gain curve.
-
In-situ does not verify the performance of the HA prescription against a specific fitting rationale to see if the HA mfg’s calculation for the prescription against that fitting rationale is calibrated accurately to it or not.
I would generalize the use of in-situ as a DIY/poor man’s verification of his HAs compared to REM being a rich man’s more complete verification of his HAs.
1 Like
As example, in my particular hearing loss, I can’t use a prescription against a specific fitting rationale like NAL-NL1 or NAL-NL2 or DSL, those specific rationale are intended to be used for the “normal” hearing loss (presbycusis) so in my case, I have to use something custom, if I use NAL-NL2 or in house* formula from Signia, then I don’t get enough mid frequencies boost, and I get much more (unneeded) high frequencies amplification, so to me, I had to do a custom fit, I had to get mid frequencies amplified with a much closer look at NAL-NL1 and reduce amplifications in the high frequencies.
So in resume, there problem is not REM, the problem (in my case) is that no formula accommodates to my cookie bite hearing loss so until a formula that fits my hearing loss, using or nor REM won’t help me to get a better hearing.
Well I think your not quite looking at this right, you can only choose one of the fitting formulas being offered in the software when programming your HAs, so if you choose NAL 1 or 2 and you then make “additional adjustments” when doing DIY, that doesn’t change the formula being used, and any formula can be used for cookie bite loss.
1 Like
Correct me if I’m wrong but the formulas available for Connexx are NAL-NL1* (I don’t know why it has an asterisk), NAL-NL2, DLS or XFit, DLS and 1/3 Gain (I don’t know what is this one), and after checking them the only ones that seem to have more adjustments are NAL (1 and 2), those have Kneepoints and ratios, the rest doesn’t.
I’m not sure if Kneepoints and ratios in Signia Connexx, are just for signia or is for all manufacturers.
So to me NAL 1 and NAL 2 seem to be the same except that NAL 2 has lower gain in mid frequencies and this are the only ones that I see different from the rest because this ones have Kneepoints and ratios.
So my question is if those are the only differences between formulas or there are other differences beside Kneepoints and ratios?
Correct me on this too but, what I’m starting to learn/discuss is that getting the exact NAL 1/2 or DLS with 100% exactitude using REM, it won’t give you the best hearing possible for cookie bite because pretty much all formulas are intended mostly for presbycusis.
Am I correct?
Yes that’s what I see.
Not quite, both a 1/3 gain and DSLv5 have adjustments for kneepoints, AXFit has it all worked out for you (I don’t know why) they just think they know best!
Adjustments for Compression kneepoints is not unique for Connexx, manufacturers can offer it, and most do.
Oh yeah there’s differences between the formula’s being offered a simple google will give you a ton of information, there’s also the compression handbook (bible) by Starkey that is a good reference on everything regarding “compression”
https://starkeypro.com/pdfs/The_Compression_Handbook.pdf
The NAL-NL2 Prescription Procedure - PMC.
Hmm, well not quite, I mean young people and children don’t have “age related loss” but regardless of all this, REM is not the holy grail of fitting, it’s certainly a useful tool for helping the audiologist’s fit up, but clearly as you have seen, it doesn’t help everyone.
When I look at your audiogram your loss is really not that bad in the “cookie bite” world, so I would think it’s not to hard to find a solution that your happy with.
1 Like
Thanks for those links, I’ve found two of those a while back but now I’ll re-read them to understand more.
I’ll recheck them because AFAIK connexx only has for NAL.
Then I’ll check the formulas for younger people, maybe those ones will fit better for me
Maybe because I’m not that old, I’m “just” 45 years old, but I’m not sure how bad will be my hearing in 10 or 20 years.
Meanwhile several of the most sold manufacturers of hearing aids have developed programs within their fitting software to do integrated REM.
For Phonak Target the integrated REM section is called Target Match. Phonak claims that the integrated REM in their Target software is done within 2 minutes (!) and with very good accuracy.
The program is based on equipment from Otometric Aurical.
Phonak also claims that using Target Match REM contributes to higher overall satisfaction and an overall better experience for clients. Sounds like a strong argument to include REM in your fitting.
More info Phonak TargetMatch Overview | PhonakPro
Also GNResound and Oticon have now integrated REM programs in their fitting software, using the same equipment.
Unfortunately the equipment from Otometric Aurical (which also is used by GN and Oticon) is quite expensive, I believe some 8-10 KUSD.
We should inform our audies about this feature and prompt them to use it.
More info about GN and Oticon REM can be found by googling on the internet
2 Likes
I once read a presentation from Audiology Online from Donald Schum (then still from Oticon) that discussed the challenges of the cookie bite loss and how Oticon did research to find out what the optimal way to correct for it, and implemented it into their VAC+ formula. In general, conventional thinking would imply that a cookie bite loss with less loss in the lows and highs but more loss in the mids would make you want to compensate for the mid frequencies’ loss by amplifying more in this region and less in the highs.
However what Oticon found is that if this were done (strong emphasis on mid frequencies amplification as a compensation) actually makes speech less understandable for people with cookie bite hearing losses. They found that more highs (not very high, but maybe in the 2-4 KHz range) would give better speech understanding for these folks. They’ve incorporated their research findings into their VAC+ fitting rationale accordingly.
So maybe you’re thinking more conventionally about this, while the HA mfgs have done more research and there’s a reason why they did what they did.
1 Like